Alibaba Qwen releases Qwen4-Max, adopting a MoE architecture and supporting a 1 million token context
Alibaba Qwen released Qwen4-Max, based on a new MoE architecture, with native vision support and a context window of 1 million tokens. The company says that in its internal evaluations, the average cost per task dropped from $4.43 to $1.67, and the Work360 score was 52.69, about 10% higher than Qwen3.8-Max. The API is now live on QwenCloud and is also available for trial in Qwen Studio, QwenWork, and Qoder, at $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching.
Why read this
AI score 76
The original text provides the new architecture, context window, and per-task cost changes, so readers can use them to judge its positioning in professional workflows.
Topics
Body
Manufacturing: from sketch to real-world industrial production. https://t.co/0BTZL8yqYJ